Introduction

For this assignment, I used Transcriptional profile of human STAT1-/- fibroblasts expressing LY6E or empty control vector data, GSE111958, 11433 genes 6 samples, which about if LY6E will enhance the infectivity of some viruses(Mar KB 2018). In assignment 1, I had already cleaned (remove the low counts) and normalized the data. The link for Assignment 1

gse <- getGEO("GSE111958",GSEMatrix=FALSE)
kable(data.frame(head(Meta(gse))), format = "html")

current_gpl <- names(GPLList(gse))[1]
current_gpl_info <- Meta(getGEO(current_gpl))

Platform title : Illumina HiSeq 2500 (Homo sapiens)
Submission data : Mar 14 2013
Last update data : Mar 27 2019
Organims : Homo sapiens
Number of GEO datasets that use this techology : 6188
Number of GEO samples that use this technology : 179516

In assignment 2, I have done differential gene expression analysis and thresholded over-representation analysis. The g:profiler result shows LY6E must play some important role in early step of the virus life cycle and infectivity of some viruses, which is consistent to the results in the paper “LY6E Mediates an Evolutionarily Conserved Enhancement of Virus Infection by Targeting a Late Entry Step.”(Mar KB 2018). The link for Assignment 1

In this assignment, I will take the ranked data that was created in Assignment 2 and perform non-thresholded pathway analysis, then summarize and visualize the results using Cytoscape and the Enrichment map pipeline.

Preparation

docker run -e PASSWORD=1234 --rm -p 8787:8787 -v /Users/bfx/Documents/BCB420:/home/rstudio/docker_bcb420 risserlin/bcb420-base-image

differentially_expressed_genes <- qlf_output_hits_withgn[which(qlf_output_hits_withgn$PValue < 0.05),]
ranked_genelist <- data.frame(differentially_expressed_genes$hgnc_symbol, differentially_expressed_genes$rank)
colnames(ranked_genelist) <- c("GeneName", "rank")
ranked_genelist <- ranked_genelist[nrow(ranked_genelist):1,]

write.table(x=ranked_genelist,
            file = file.path("data", "ranked_genelist.rnk"),sep = "\t",
            row.names = FALSE,col.names = TRUE, quote = FALSE)

Non-thresholded Gene set Enrichment Analysis

Visualize Gene set Enrichment Analysis in Cytoscape

Use Cytoscape 3.7.2 to create the Enrichment Map by following the tutorial

• input the GSEA result
• Set FDR q-value cutoff to 0.1
• Build

The map is very large and no red node, which caused by that there is no hit has FDR q-val < 0.1 in upregulate. So I select show advanced options and set q-value to 0.5, p-value to 0.0001 to reduce the number of nodes and keep enough red node. The enrichment map has 285 nodes and 682 edges

Annotate the network by AutoAnotate and label column with GS_DESCR

remove the clusters which has less than 6 nodes

manually arrange the nodes to avoid the overlap

The legend of the enrichment map

Collapse the network, there are 8 theme:

• glycolytic process pyruvate
• collagen surface extracellular
• electron mitochondrial respiratory (upregulate)
• substrate adherens junction
• epiboly wound healing
• body fluid coagulation
• organ growth bone
• eukaryotic translation elongation

They fit with the model Since the original paper conclude that LY6E promotes early step of the virus life cycle and enhances infectivity of some viruses and mitochondria is a platform for antiviral innate immunity and antiviral innate immunity requires oxidative phosphorylation (OXPHOS) activity.(Yoshizumi 2017). And also the original paper said the LY6E will not influence the host transcriptome, for these 8 themes, only one is upregulated which also suggest that LY6E may affect the infectivity directly.

Interpretation and detailed view of results

For the enrichment results, Most are downregulated. Same with the conclusion of Assignment 2, the mitochondria related theme is upregulated which support the conclusions discussed in the original paper that LY6E promotes early step of the virus life cycle and enhances infectivity of some viruses. Also I found an article “RLR-mediated antiviral innate immunity requires oxidative phosphorylation activity”, published at 2017, said mitochondria is a platform for antiviral innate immunity and antiviral innate immunity requires oxidative phosphorylation (OXPHOS) activity.(Yoshizumi 2017), the downregulate themes seems like also facilitate the infectivity of viruses. Therefore LY6E must play some important role in early step of the virus life cycle and infectivity of some viruses.